Increased levels of hydrogen peroxide (HOOH) is thought to occur under conditions of tissue reperfusion with oxygenated media after hypoxia. Since the vascular endothelium is the first in line of exposure of circulating hemoglobin-based blood substitutes, we investigated the effects of a variety of modified hemoglobins on the integrity of bovine aorta endothelial cells (BAEC) and on the HOOH mediated cytotoxicity. We found that unmodified hemoglobin AO to be protective against peroxide- induced endothelial cell necrosis, probably due to its peroxidase activity (hemoglobin consumes HOOH as its cycles between the ferric and the ferryl hemes), and that this pseudo-enzymatic activity is dependent on the nature of structural modifications of the protein. Intra-and intermolecular cross-linked hemoglobin, however, did not protect against peroxide-induced endothelial cell damage. Chemically modified hemoglobins appear to have an inherent ability to produce a persistent ferryl species. Using a model of ischemia and reperfusion (I/R) which uses endothelial cells aggregated on microcarrier beads, we demonstrated that cell survival after I/R was not significantly decreased by doses of cross-linked hemoglobin less than 100 muM. At these doses, however, hemoglobin decreased the level of hydrogen peroxide in the media and increased lipid peroxidation of both normal and I/R cells, an effect which may be expected to lead to more subtle forms of toxicity. Hemoglobin itself was oxidized to some degree by normal cells and to a greater extent by ischemic cells. After ischemia, ferryl hemoglobin was transiently formed in the first few hours of reperfusion, during a time frame in which lipid peroxidation is also at its peak. Site directed mutagenesis is a potentially effective tool for the engineering hemoglobin because it allows the fine tuning of protein function and stability. At this point, myoglobin has provided a simple prototype for these experiments, and that indeed a number of myoglobin mutants have been prepared in collaboration with Rice University that are have different ligand binding, autoxidation and stability. We have data to show clear differences among these myoglobins in terms of the rates of ferryl formation and its persistence in solutions and that selectively constructed mutants will be used to probing the mechanism of HOOH mode of entry into the proteins. This will ultimately help in the design of a protein that can stereochemically restrict the entry of HOOH and minimize its oxidative effect.